A Novel Real-Time Detection and Classification Method for ECG Signal Images Based on Deep Learning

Area of Science:

• Medical Imaging and Diagnostics
• Artificial Intelligence in Healthcare
• Signal Processing

Background:

• Electrocardiogram (ECG) analysis is crucial for diagnosing cardiac conditions.
• Current methods for ECG image processing face challenges in real-time accuracy and efficiency.
• Deep learning offers potential for improving automated ECG interpretation.

Purpose of the Study:

• To introduce Mamba-RAYOLO, a novel deep learning method for enhanced ECG image processing.
• To improve the real-time detection and classification accuracy of ECG images.
• To provide a more efficient framework for medical ECG diagnostics.

Main Methods:

• Developed Mamba-RAYOLO, integrating feature extraction, attention mechanism, and feature refinement modules.
• Employed a multi-branch structure for comprehensive feature extraction during training.
• Utilized an attention mechanism for dynamic focus on relevant spatial and channel-wise features.

Main Results:

• Mamba-RAYOLO demonstrated significant improvements in ECG image detection and classification.
• The method achieved enhanced accuracy and computational efficiency in experimental tests.
• Feature extraction and fusion modules contributed to robust performance.

Conclusions:

• The proposed Mamba-RAYOLO method offers a promising advancement in ECG diagnostics.
• The integrated modules enhance the precision and speed of ECG image analysis.
• This framework supports more accurate and efficient medical diagnoses.